Hacker News Reader

Comment by nine_k | Hacker News Reader

It's not even an Adobe fault likely. It's Pantone's market segmentation, AFAICT. "We don't want cheapskates who would ever mind paying mere $21/mo for access to Pantone™ colors", they could say. They want to deal with larger print shops / agencies, and better monetize their chokehold on the industry-accepted color-matching system.

This makes a niche for a more open, non-commercial color-matching system for print. Which has few incentives to happen: print color is hard, buy-in from paint producers is likely even harder.

For Web and other screen-only work, I suppose you don't need Pantone.

View on HN

That niche is filled pretty nicely by Heidelberg. Their printers come with their own color correction and matching system. Plus they ship ICC profiles for accurate screen preview.

I was looking at Heidelberg since you suggested it. But, their product catalog is barely visible outside of spare parts and inks. The few things I could see, I cannot use them in a print shop, their equipment is much larger than the current print equipment which is nearing end of life.

> This makes a niche for a more open, non-commercial color-matching system for print.

Wikipedia tells me Pantone has been around since 1950, TL;DR if open non-commercial was going to happen, it would have happened already.

As you say, few incentives. Its harder than it looks, its not just a case of someone randomly declaring FOO 123=CMYK XYZ. Further, the whole supply chain is already well established with Pantone.

We have CIELAB for expressing colour in absolute form. This could always be used to express an intended colour.

However you are correct Pantone is entrenched as it is based on a simple set of formulas for easy mixing by print shops, this is also cemented into various systems that take premixed pantones directly (eg HP Indigo). Pantone also extends into neon and metallics where LAB of course doesn’t exist.

Declaring arbitrary CMYK or RGB values without including a corresponding colour profile will produce inaccurate results across different devices/output methods due to variation (hence why we have colour profiles.)

This change will just likely lead to more colours being specified as arbitrary rgb and cmyk values with accuracy be damned, which is a trend of new style guides for the last few years (also due the rise of trademarking colours).

I know of at least one color lab that spent years and had multiple really smart people trying to come up with a closed-form solution for the color (both diffuse and specular) given a {ink, illuminant, substrate} combination and last I heard never succeeded. Without that, you are limited to making giant tables, which is what PMS does and it's only useful if everyone is using the same tables.

I think part of the reason is that beyond a certain scope there is limited commercial desire for such levels of exactness.

While companies will talk the talk about always needing their brand colour represented perfectly. I've found very few walk the talk. Presenting printing quotations with PMS versus standard CMYK, they often compromise. Presenting premiums or other production which isn't print-based, suddenly they're happy to compromise rather than organise a custom dip.

So while we have CIE as our solid reference, for cosmetic representations, we probably don't need more beyond than that commercially for the bulk of clients.

How you even define metallic paint in numbers ? Both shape and type of particles can differ

Bit of a cheeky response but you could always encode the BRDF with numbers.

Reply on news.ycombinator.com